Funded by the European Union

COVID-19 SNAPSHOT in December 2020 Conflict Introduction The ongoing conflict in Yemen is complex and multi-layered, with the involvement of rival regional powers adding a geopolitical dimension to the war. As the national framework has disintegrated, local rivalries have intensified, leaving more room for foreign state interference in the country.1 Since March 2015, a dangerous combination of factors, driven by conflict and economic decline, and now exacerbated by COVID-19, have compounded the dire situation for over 24 million Yemenis, as reported by the UN, international non-governmental organisations (INGOs), and local sources. The country has been de facto split into two since several years: the North, controlled by the de-facto authorities of Ansarullah “Al-Houthi”, and the South, which is controlled by the internationally-recognised Hadi government (IRG).

Data collection and analysis Data for this first COVID-19 snapshot report in Yemen Conflict was gathered from 31 October to 10 November 2020 in the following target zones: Tuban and Toor Al-Baha districts in Lahj governorate (located in the South, under IRG control), and Al-Mawasit and Jabal Habashi districts in governorate (located in the North but under IRG control). This report was prepared by an independent consultant and the Search for Common Ground (Search) Yemen team, with support from Search’s COVID-19 Response Programming Team and Institutional Learning Team. For more information on the data collection methodology and tools used by Search for Common Ground for this research, click here.

To cite this report: Search for Common Ground, COVID-19 Snapshot in Yemen Conflict, December 2020.

* This publication was produced with the financial support of the European Union. Its contents are the sole responsibility of Search for Common Ground and do not necessarily reflect the views of the European Union.

1. Ardemagni, Eleonora, Beyond Yemen’s Militiadoms – Restarting from local agencies, European Union Institute for Security Studies (EUISS), April 2020. Update: COVID-19 Measures and Impact on Conflict

As of 6 December 2020, the number of confirmed COVID-19 cases in Yemen reached 2,081, with 607 associated deaths. The number of confirmed cases is very likely to be (much) lower than the number of actual cases given limited testing.2 During the past three months, no particular COVID-19 prevention measures were in place in the North and South of Yemen: As the Yemeni civilian population is already suffering from an enormous man-made humanitarian crisis, the COVID-19 crisis has added grave new stresses to a health system already shattered by war. Under-resourced and buffeted by years of conflict, Yemen’s health system is inadequately prepared to care for COVID-19 patients and contain the spread of the virus. Displaced populations in particular face additional challenges; the pandemic and lockdown measures have heightened protection concerns and resulted in significant health and safety risks.3 Many households lack access to healthcare and require information about COVID-19, including on health and safety measures, symptoms of COVID-19, what to do when COVID-19 symptoms emerge, and the guidelines and protocols enacted by authorities. The lockdown measures have increased food and economic insecurity and exacerbated the struggle of vulnerable households to access basic services, in camp and non-camp settings alike.

The pandemic health and safety guidelines call for restricted movement and social isolation measures, which have consequently led to increased tensions at the household level and higher rates of domestic violence. Specifically, women and girls are forced to stay in crowded households and have limited accessto information and support services. This situation increases their level of vulnerability as well as risks of domestic abuse. The rate of violence against women in Yemen was already very high in the context of the ongoing conflict – in 2017, UNFPA reported that 2.6 million women and girls were at risk of gender-based violence.4 With the added economic, health, and social stressors of COVID-19, domestic violence cases are on the rise.5

COVID-19’s Impact on Social Cohesion & Trust Trust in COVID-19 information from Search’s media partner is relatively low, and information sharing is limited know Search’s local radio partner, Sawt Al-Tofola FM, except in 62 2 Al-Mawasit district55 in , 2 2 55 55 where none of the respondents reported knowing our radio partner trust (very much lower trust do not or mostly) (somewhat or rarely) trust it

Trends Young people (18-34 years) tend to trust information on COVID-19 to shared through our local radio partner more than respondents over 56 6 Follow 13 3 35 years.

There was no marked difference in trust between IDPs and host communities, indicating this is a good radio channel to reach both groups with COVID-19 information.

Overall, 33% of survey respondents shared COVID-19 related information 67 heard on Sawt Al-Tofola FM Radio with their friends and family members.

Trends On average, young respondents between 25 to 34 years reported a to 7 Follow higher level of information sharing, compared to the other groups (18 to 3 1 24 and above 35 years). 3

The degree of information sharing on COVID-19 with friends and family varies significantly across Taiz districts. Lahj 59% Jabal Habashi 68% Toor Al-Baha 4% Al-Mawasit 5% Tuban

2. Our World in Data Portal, November 2020. 3. Norwegian Refugee Council, COVID-19 Impact Assessment, July 2020. 4. United Nations Population Fund, UNFPA Humanitarian Response in Yemen, October 2017. 2 5. UN WOMEN, In Yemen, women face added challenges posed by COVID-19 amidst ongoing armed conflict, July 2020. One out of two are satisfied with COVID-19 services, a slight majority trust the government’s pandemic response, more people have faith in COVID-19 responders - all with important differences between districts.

72% 8% 2 health access to food

sought out services or support sought support The top requested services requested were related to COVID-19 in the past 3 from the months, despite high levels of need. government

Satisfaction with services received Satisfaction with the speed of service Highly satisfied Highly satisfied and satisfied 55 and satisfied Somewhat Somewhat 2 satisfied satisfied

Not satisfied 2 Not satisfied 22

Never received Preferred not a response 2 to answer 2

Women are more satisfied with the services received and Trends to the response time than men. This might be attributed to the fact Follow that women on average may have lower expectations than men, or that female-headed households may have benefited more from specific COVID-19 emergency response, compared to other 72 2 6 2 households.

Younger people (18 to 24 years) reported the lowest level of satisfaction with services and the response time. This could perhaps be attributed to 1 o 25 differences of expectations and aspirations among younger and older people.

IDPs reported higher levels of satisfaction than host community members Satisfaction 67 with 5 with services provided. Often, emergency aid interventions in Yemen tend to 67 services 5 received focus more on IDPs compared to host community groups, which could explain o this difference. 62 o on on There are differences in satisfaction between districts: the lowest level of satisfaction was reported in (Taiz), followed by Toor Al-Baha (Lahj), Tuban (Lahj), and Taiz Al-Mawasit (Taiz). Lahj 21% Jabal Habashi 53% Toor Al-Baha 71% Al-Mawasit 67% Tuban

agree (strongly agree or agree) that the authorities, including 5 institutions2 and service2 providers (public and private), are 5doing5 22 22 their best to consider the needs of everyone equally when making decisions about COVID-19 services. disagree neutral

Trends Women show higher levels of trust than men in the authorities’ to Follow response. 72 6

Levels of trust in the government’s response tend to be higher among youth (18 to 34 year olds) and lower among respondents above 35 years old. While levels of satisfaction with services among young people are lower, as noted above, 7 5 their trust in the government remains high. This may reflect the notion that while 13 3 services might not always be satisfactory, youth still trust that the government is doing its best given the prevailing context. Whereas older respondents’ lower levels of trust in government might be due to their pre-existing, historical relationships with and mistrust of government over time, while their higher satisfaction in services could potentially reflect the lower expectations they have in services in general. 3 Trends Average levels of trust in the authorities’ COVID-19 response differs between districts.A large to Follow majority of respondents in Toor Al-Baha district (Lahj) agree that the government is doing its best, followed by Jabal Habashi district (Taiz), Al-Mawasit district (Taiz) and Tuban district (Lahj). 58% Jabal Habashi Taiz 54% Al-Mawasit Lahj 84% Toor Al-Baha 43% Tuban

As in other countries under this research, trust in COVID-19 responders tends to be higher than in the government, although this difference was rather small in Yemen. agree (strongly agree or agree) that disagree people working on COVID-19 have (strongly 6 6 6 neutral their best interests in mind when disagree or they do their work disagree)

Overall, 82% of women agree with the above statement, compared Trends to to 56% of men. One reason behind this difference could be that women Follow 2 56 may have more regular contact with health workers than men, but more research is needed to better understand this gender gap.

Trust in COVID-19 responders is higher among young people (ages 18 to 34) compared to those above 35 years. This may be attributed to the fact that young 62 people generally have greater access to information and awareness on COVID-19 13 3 (i.e. via social media) and thus better understand the risks facing health workers and other responders, which may lead to higher appreciation for their work. IDPs tend to mistrust people working on COVID-19 more than host community members. Certain vulnerabilities, barriers, and realities that displaced populations 25 face, might explain greater levels of mistrust. For instance, negative experiences with other forms of humanitarian response interventions could translate into wider o levels of mistrust towards COVID-19 responders. on

Mistrust towards COVID-19 responders varies from one district to another. Mistrust is the highest in Al-Mawasit district (Taiz), compared to Toor Al Baha, Tuban (Lahj) and Jabal Habashi (Taiz). Levels of mistrust could be attributed to the real or perceived Lahj Taiz absence of health workers at the beginning of the 16% Jabal Habashi 21% Toor Al-Baha Ibb pandemic outbreak in Aden city, limiting citizens’ 34% Al-Mawasit 21% Tuban ability to access COVID-19 related services.

Despite years of conflict, a majority of Yemenis support collaboration across dividing lines to address a crisis like COVID-19, but opportunities for actual interaction are limited General trends about collaboration and interaction across groups send important signals for conflict. However, when examining horizontal cohesion, the most important trends to pay attention to are the outliers: who values this type of collaboration, who does not, and what influences willingness to collaborate across different dividing lines. The majority of respondents in this study see the value of collaboration across divides to address a crisis like COVID-19. This includes collaboration between people from areas under different political control, as well as collaboration between IDPs and host communities.6

One of the main dividing lines in Yemen is between IDPs and host communities. Often relationships between IDPs and host communities are very tense, due to a common perception among host communities that humanitarian organisations focus disproportionately on IDPs, resulting in community-level conflicts across affected districts. Thus, prioritising IDPs for COVID-19 services - given the context of a protracted humanitarian crisis as well as the fluidity of the term ‘IDP’ within the Yemeni context - is likely to increase tension and decrease interaction among these communities.

7 see the value of collaboration 7 7 between host communities and disagree neutral IDPs 4 Women and youth value collaboration between IDPs and Trends to host communities more than men and adults. The majority Follow 6 7 of COVID-19 community volunteers in Yemen are from youth 13 3 groups, male and female. They are leading efforts to produce PPE for healthcare workers and people affected by COVID-19. Yemeni women are taking the lead at the local level in the fight against COVID-19, and advocate for the importance of collaboration among different community segments to respond to the COVID-19 crisis.

Perhaps not surprisingly, IDPs value this kind of collaboration more than the host communities, as host communities tend to have more hostile feelings 2 7 towards IDPs than the other way around. o on

Another major dividing line in Yemen is between different political entities and their respective supporters. Our study found that see the value of collaboration across 7 7 7 no political dividing lines to address a disagree opinion crisis like COVID-19

Women and youth value collaboration across political Trends to dividing lines much more than men and adults.This may 2 72 Follow 62 be because, generally, women and young people are often 13 3 excluded from directly participating in political life in Yemen and therefore may not experience political divisions as acutely as men and adults. There is also a clear difference betweenIDPs and host communities: 100% of IDPs see the value in collaboration across political dividing lines compared to 79% of the 7 host community. This may be because IDPs are more vulnerable to the direct repercussions resulting from the lack of collaboration between political entities. o on

Moreover, the degree to which collaboration across political dividing lines is valued differs between districts:the vast majority

of respondents from Jabal Habashi district (Taiz) value this Taiz collaboration, followed by Al-Mawasit (Taiz), Tuban 94% Jabal Habashi Lahj 69% Toor Al-Baha and Toor Al-Baha (both in Lahj). 79% Al-Mawasit 72% Tuban

Moreover, we found that horizontal cohesion between people from different regions within Yemen remains high, despite years of ongoing conflict: 2 value collaboration with people2 2 from other districts in responding disagree neutral to a crisis like COVID-19

Again, young people and women There is a slight difference between IDPs and host Trends to value this type of collaboration communities, with IDPs more likely to agree that this Follow more: collaboration is important compared to host community members. host 6 67 5 2 community 13 3 IDP o on

6. Over the past years, many Yemenis have migrated within the country (e.g. from Aden, Taiz, or other governorates to Sana’a). More recently, thousands of IDPs have settled in the governorate of Marib, in many cases finding refuge with their relatives or friends. Therefore, pre-existing high levels of co-existence between 5 and among different groups and regions in Yemen could translate into higher levels of solidarity during COVID-19, particularly within the extended family. Our research also seeks to understand to what extent groups and individuals across dividing lines are still interacting throughout the pandemic. To start, respondents were asked how often they interacted (in-person or virtually) with individuals and/or groups across dividing lines during the past 3 months of the COVID-19 crisis and lockdown.7 We will track this indicator over time to assess how COVID-19 is impacting opportunities for people and groups to interact across divides, both for daily interactions as well as to mediate conflicts.

5 More than5 half of respondents from IDP groups and host 5 5 communities interacted (remotely or in-person) with people from the other group in the last 3 months. o on o on

Unsurprisingly, men from both groups interacted more Trends to frequently than women. And more youth (18 to 34 years) Follow have interactions with people from the other group than 62 5 people above 35 years. This may be explained by the fact 1 3 3 that younger people are more active on social media.

5 About half of survey2 respondents5 2 had no interacted with individuals from interaction other districts in the past 3 months

Men interact with people from other districts more frequently Trends 6 5 to than women. Respondents aged 18 to 24 reported the lowest 6 Follow level of interaction compared to older respondents aged 25 and above. 1 There are variations at district level,

Taiz 45% Jabal Habashi Lahj 51% Toor Al-Baha 57% Al-Mawasit 49% Tuban

Next Steps For COVID-19 Response Efforts Prevent increasing tensions between IDPs and host communities by providing integrated COVID-19 services • While social cohesion and peaceful coexistence between host and IDP communities remain relatively high, providing COVID-19 support services separately to IDPs and host communities risks aggravating local tensions, as the support would be perceived as benefiting only one side. This pre-existing dynamic should be taken into account by authorities, public health responders and humanitarian actors to ensure that the COVID-19 response does not do harm and integrates conflict mitigation measures.

Offer more opportunities for (virtual and in-person) interaction across dividing lines to strengthen social cohesion during the pandemic • Overall, we found a majority of people support collaboration across dividing lines to address a crisis like COVID-19, and this despite the years of ongoing conflict in Yemen. However, people have limited opportunities for actual interaction with individuals and groups who are different from them. Especially certain population groups like women, host community youth, and IDP adults experience low rates of interaction. These groups would benefit from increased opportunities to interact, virtually or in-person, where it is safe to do so. This finding also indicates a unique opportunity for the pandemic response to contribute to larger social cohesion and peaceful coexistence goals in Yemen, if done the right way. Public health responders and peacebuilders should work together to better understand these dynamics and to reinforce cohesion and unity across communities through pandemic response interventions.

6 7. Measured on a daily, weekly, monthly, or once per quarter basis.